vaayaa Inclusive Education

Home › Micro Degree Best Practices › Search Engine Marketing and Paid Search

Best Practice Manual

Search Engine Marketing and Paid Search

A practical reference for planning, launching and refining paid search campaigns, from keyword research to conversion tracking.

Marketing · By vaayaa Editorial Team · Updated Oct 30, 2026

01Purpose and scope

  • This manual supports the course Search Engine Marketing and Paid Search. It describes how paid search advertising is planned, built, run and reviewed in day-to-day practice, so that a practitioner can use it as a working reference rather than a theory text. The focus is on advertisements that appear beside or above organic results when a person types or speaks a query, and on the paid placements that appear on partner sites in the same auction-driven format.
  • The manual covers small local accounts, mid-sized e-commerce catalogs and larger multi-market programs. Practice changes a great deal between these settings. A local service business may run three campaigns with a daily budget that one analyst can review in ten minutes, while a retailer may manage tens of thousands of keywords across dozens of product feeds and several languages. Where the approach differs, the text says so.
  • Paid search sits inside a broader search engine marketing function that also includes organic search work. This manual does not teach search engine optimization, display advertising, social advertising or email. It touches those areas only where they affect search budgets or measurement. Legal and policy requirements for advertising vary by country, state and product category, and the current rules should always be confirmed with the relevant ad platform policy pages and the authority that regulates the advertised product.

02Core principles

  • Paid search is an auction. Each time a query triggers eligible ads, the platform ranks advertisers using a combination of the bid, the expected likelihood of a click, the quality of the landing page and the relevance of the ad to the query. Money does not buy a fixed position. A practitioner who understands this can explain why an account with a lower maximum bid sometimes wins more impressions than a competitor with a higher one.
  • Intent drives value. A query such as emergency plumber near me signals a different buying moment than best plumbing tools for beginners. Good campaigns group keywords by intent, write ads that answer the specific need and send visitors to pages that continue the same thought. Mixing intents in one ad group usually weakens relevance and makes bids harder to set.
  • Measurement must match the business goal. Clicks and impressions are inputs, not outcomes. The core outcomes are usually qualified leads, completed purchases, booked appointments or revenue per visit, and each needs a tracking method that the business trusts. If the tracked conversion does not reflect what the business values, optimization will drift toward the wrong result.
  • Control is partial. Platforms use automated bidding and ad delivery systems that learn from historical data. The practitioner sets goals, budgets, constraints, audiences and creative, then reviews whether the system behaves as intended. Automation saves labor in mature accounts with reliable conversion data and can cause expensive mistakes in accounts without it.

03Planning and preparation

  • Start with the offer, the margin and the capacity to serve demand. Before any keyword is chosen, confirm what the business can sell or deliver, what profit it earns per unit or per lead, and how many new customers the sales team can handle in a week. Paid search that brings in more demand than the operation can serve wastes money and damages reputation. Ask for a clear target cost per acquisition or return on ad spend, and write down the assumption behind it.
  • Audit the existing account and site before building anything new. Check conversion tracking, landing page speed, mobile usability, product availability, pricing accuracy and any disapproved ads. Review the search terms report from the last few months to see which queries already trigger ads and whether they match the intended offer. Many accounts that appear to need new campaigns actually need cleaning.
  • Build a keyword plan from several sources: the platform keyword planning tool, the site's own search logs, customer support transcripts, sales call notes and competitor ad copy visible in public results. Treat volume estimates as rough guides. Use them to rank opportunities, not to forecast revenue precisely.
  • Decide the account structure before launch. Campaigns separate budgets and settings such as location, language, bidding strategy and network placement. Ad groups separate themes. A common pattern is one campaign per product line or service, with ad groups organized by tight keyword themes that share a landing page. Changing structure later is possible, but it resets learning and makes historical comparisons harder.
  • Set up tracking and budget rules first. Install the platform conversion tag or import offline conversions from the customer relationship system. Confirm that a test submission registers correctly. Define the daily budget per campaign, the maximum cost per click for manual bidding if used, and the threshold at which a campaign will be paused or reduced.

04Recommended workflow

  1. 1
    : Define one measurable objective for each campaign, such as cost per qualified lead under a stated amount or purchases at or above a stated return on ad spend. Record the reason for the target so that later reviewers understand the trade-off being made.
  2. 2
    : Research keywords and group them by intent, product and landing page. Use match types deliberately. Broad match reaches more queries but needs strong negative keyword lists and reliable conversion data. Phrase and exact match give tighter control in newer accounts.
  3. 3
    : Write responsive or standard ad copy for each ad group. Include the core term in at least one headline, state a concrete benefit or price, add a clear call to action and use extensions such as sitelinks, call details or location where they fit the business. Avoid claims the landing page cannot support.
  4. 4
    : Build or choose the landing page. The page should repeat the promise in the ad, show the offer above the fold, load quickly on mobile and make the next action obvious. Route each ad group to the most relevant page rather than sending all traffic to the homepage.
  5. 5
    : Set bids and budgets. Start with manual or target-based bidding if conversion volume is low. Use a daily budget that allows the campaign to collect enough data to learn, and avoid changing targets more than once every week or two unless something is clearly broken.
  6. 6
    : Launch, then review the first search terms report within the first several days. Add irrelevant queries as negative keywords at the right level, either the ad group or the campaign. Pause keywords that spend without converting after a fair test window agreed in advance.
  7. 7
    : Review performance on a fixed cycle. Compare cost, conversions, conversion value and impression share against the target. Make a small number of changes at a time so the effect of each is traceable. Record every change with the date and reason in the account change log or internal notes.

05Quality checks, risk controls and compliance

  • Check ads before they go live. Confirm that every headline and description matches the landing page, that prices and shipping terms are current, that trademarked brand names are used only as the platform permits, and that the destination URL resolves without errors. Run a preview across devices and locations if location targeting is narrow.
  • Watch the account for policy issues. Platforms restrict or prohibit certain categories, including some financial products, health claims, gambling, weapons, adult content and political advertising. Many categories also require advertiser verification or a license. The platform policy center is the authoritative source for what is allowed and what documentation is needed. Do not rely on a summary written by someone else.
  • Protect privacy and consent. Tracking pixels, audience lists and remarketing tags may trigger consent requirements under privacy laws that differ by region. Work with the legal or privacy team to confirm which tags fire before consent, how consent signals are passed to the platform and how customer data is hashed or uploaded. Requirements change, so confirm the current rules with the relevant regulator or counsel.
  • Control spend risk. Set account-level spending limits where the platform allows them. Review automated bidding changes after any large budget increase. Watch for click fraud signals such as unusually high clicks from a single location, very low time on site or sudden traffic from unrelated regions. Report suspected invalid activity through the platform and keep a record.
  • Regulated or high-risk offers need extra care. If the advertised product involves medicine, legal services, financial products, tax advice or safety-critical goods, this guidance does not replace qualified professional judgment, approved procedures or any required authorization. Compliance review by the responsible specialist should happen before launch, and the copy should not promise outcomes the business cannot guarantee.

06Collaboration and documentation

  • Paid search rarely succeeds alone. The practitioner depends on the web team for landing pages and tag installation, the product team for feed accuracy and stock status, sales for lead quality feedback, finance for margin and budget approval, and legal or compliance for claims and privacy. Agree on who owns each input and how long a request normally takes. A two-day delay for a landing page change can cost a week of learning in a small account.
  • Keep a written account brief. It should state the business goal, target metrics, budget, campaign structure, naming convention, approved claims, excluded terms, contact persons and the review schedule. A new team member should be able to read it and understand why the account looks the way it does.
  • Maintain a change log. Each entry should record the date, the person making the change, what changed, the reason and the expected effect. When performance shifts, the log lets the team separate an algorithm change, a seasonal demand shift, a competitor entry and an internal edit.
  • Use consistent naming for campaigns, ad groups, labels and tracking parameters. Apply UTM parameters or equivalent tags to destination URLs so that web analytics can attribute visits accurately. Store a copy of the feed rules, negative keyword lists and audience definitions in a shared location rather than in one person's inbox.

07Measures and continuous improvement

  • Track a short set of indicators. Useful measures include cost per click, click-through rate, conversion rate, cost per conversion, return on ad spend, impression share lost to budget and lost to rank, search term match rate and the share of spend on negative or irrelevant queries. Look at them together. A falling cost per click that comes with a falling conversion rate is not an improvement.
  • Separate leading and lagging indicators. Click-through rate and quality indicators show whether the ad is relevant early. Revenue, qualified lead rate and customer lifetime value show whether the campaign pays off later. For businesses with long sales cycles, use offline conversion imports so the platform learns from leads that became customers rather than from form fills alone.
  • Test with discipline. Change one major variable at a time, such as ad copy, landing page, bid strategy or audience layer. Run the test long enough to gather meaningful conversion counts. Small accounts may not reach statistical confidence on conversion rate differences, so judge changes with the cost and value context in mind instead of chasing small percentage moves.
  • Review the market context. Auction prices change with competitor activity, seasonality, economic conditions and platform policy updates. A rise in cost per click does not always mean the campaign got worse. Compare against category benchmarks drawn from the business's own history rather than from generic industry averages, which are often unreliable.

08Common mistakes and how to prevent them

  • Buying broad keywords without negatives is the most common cost leak. Prevent it by reviewing the search terms report on a fixed schedule and adding negatives at the right level. Build a shared negative list for terms such as free, jobs, careers and definitions when those do not match the offer.
  • Sending all traffic to the homepage lowers relevance and conversion rate. Map ad groups to specific landing pages before launch and check that the page content matches the ad promise.
  • Trusting platform conversion counts without checking them against the customer system leads to inflated results. Reconcile a sample of conversions each month, check for duplicate tags and confirm that form submissions are counted once.
  • Changing bids, budgets or structure too often prevents learning. Set a review cadence and avoid reacting to a single day of data. Note that automated strategies need time and a steady volume of conversions before they settle.
  • Ignoring the sales feedback loop wastes budget on leads that never close. Ask sales to mark lead quality in the customer system and feed that back as offline conversions or as a lead score.

09Practical checklist

  • Confirm the business objective, target cost per acquisition or return on ad spend, and the margin behind it.
  • Verify conversion tracking with a test submission and reconcile against the customer or sales system.
  • Audit the existing account, search terms, landing pages and disapproved ads before building new campaigns.
  • Group keywords by intent and choose match types that fit the account's data maturity.
  • Write ads that match the landing page, state supportable claims and include relevant extensions.
  • Check policy and verification requirements for the product category in the platform policy center.
  • Confirm consent and privacy requirements with the responsible team before enabling remarketing tags.
  • Set daily budgets and account spending limits, and record the threshold for pausing a campaign.
  • Review the search terms report early and regularly, and add negative keywords at the correct level.
  • Log every change with date, owner, reason and expected effect.
  • Review cost, conversion value and impression share against target on a fixed schedule.
  • Confirm that regulated or high-risk offers have qualified professional review before launch.

★Learning sources

?Frequently asked questions

How long should a new paid search campaign run before judging it?

Most practitioners wait until the campaign has collected enough conversions to show a pattern, which usually means several weeks for a modest budget and far less for a high-volume account. Agree on the minimum test window and the spend threshold before launch so the decision is not made in reaction to a single slow day. Early signals such as click-through rate and relevant search terms can guide small fixes while the test runs.

Should I use automated bidding from the start?

Automated bidding works best when the account has steady, accurate conversion data, so new accounts often begin with manual or simpler target-based bids. Switching too early can leave the system learning on sparse or inaccurate data. Move to automation once the conversion counts are reliable and the target is realistic for the budget.

How do I handle a sudden jump in cost per click?

Check first whether the change is limited to one campaign, device, location or query group, and look for a competitor entry, seasonal demand or a bid strategy change in the change log. Compare the jump against conversion rate and cost per conversion before deciding whether the campaign got worse. If results are falling, reduce bids or budget temporarily and review the search terms for new irrelevant queries.

Who is responsible for policy and privacy compliance in paid search?

The practitioner normally handles day-to-day policy checks, but final responsibility usually sits with the business owner and, where applicable, the legal or privacy team. Requirements vary by country, state and product category, so consult the current platform policy pages and the relevant regulator or counsel. This guidance does not replace qualified professional judgment or any required authorization.

This manual explains practice in general terms. It does not replace qualified professional judgment, approved procedures or required authorization.

Marketing Best Practices

China, India, United States, Indonesia, Pakistan, Nigeria, Brazil, Bangladesh, Russia, Japan, Mexico, Philippines, Vietnam, Egypt, Turkey, Germany, United Kingdom, France, Italy, Canada, South Korea, Australia, Spain

Hourly Active: 48,359 Learners